lable at ScienceDirect

Journal of Archaeological Science 50 (2014) 369e382

Contents lists avai

Journal of Archaeological Science

journal homepage: http : / /www.elsevier .com/locate/ jas

The chronology of megalithic funerary practices: a Bayesian approachto Grave 11 At El Barranquete necropolis (Almería, Spain)

Gonzalo Aranda Jim�enez*, Agueda Lozano MedinaDepartment of Prehistory and Archaeology, University of Granada, Campus Cartuja s/n, 18071 Granada, Spain

a r t i c l e i n f o

Article history:Received 7 May 2014Received in revised form16 July 2014Accepted 1 August 2014Available online 12 August 2014

Keywords:Radiocarbon datingBayesian statisticFunerary ritualMegalithic phenomenonCopper AgeBronze Age

* Corresponding author.E-mail addresses: garanda@ugr.es (G. Aranda Ji

gmail.com (A. Lozano Medina).

http://dx.doi.org/10.1016/j.jas.2014.08.0050305-4403/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

For the first time on the southern Iberian Peninsula it is possible to determine the timescale and funeraryspan of a single megalithic grave, as all the Minimum Number of Individuals identified by anthropo-logical study have been dated. Thirteen radiocarbon measurements are now available from Grave 11 at ElBarranquete necropolis. Two Bayesian models have been built on the basis of archaeological in-terpretations of the mortuary depositions. The results stress the late construction of the monument,probably in 2452e2316 cal BC, and the short, but intensive ritual use during the Chalcolithic period ofbetween three and nine generations. The funerary reuse of the monument is one the most remarkablefeatures of this tomb. According to the Bayesian models, these ritual practices began in 2154e2022 cal BCand spans a long period of at least half a millennium. The results are also discussed in the context of themegalithic phenomenon on the southern Iberian Peninsula.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

The megalithic necropolis of El Barranquete lies in a plains areato the west of a seasonal river, the Rambla de Morales, near theMediterranean coast (c. 7 km as the crow flies) in the present-dayprovince of Almería (Fig. 1). Discovered in 1968, the necropoliswas immediately excavated by Manuel Fern�andez Miranda andMaría Josefa Almagro Gorbea during different fieldwork seasonsbetween 1968 and 1971. In 1973, a full report on all the excavationsundertaken during this period was published by María JosefaAlmagro Gorbea (1973). This report also included an anthropolog-ical study of the first eight burials undertaken by Miguel BotellaL�opez. The investigations revealed at least 17 megalithic graves, ofwhich 11 were excavated (Fig. 2). All burials conform to the clas-sification of the megalithic type known as a tholoi or tombs withchambers covered by false vaults. The chambers were enteredthrough passages that were normally divided into equal segmentsby slabs with holes in them. Small side chambers in both the pas-sages and the main chambers were also a common feature. Thesetombs were covered by mounds built with concentric stone wallsfilled with earth and small stones (Almagro Gorbea, 1973).

m�enez), aguedalozanomed@

Grave 11 is a perfect example of this type of megalithic grave(Fig. 3). The structure itself is defined by a circular chamber withdry stone walling and is 3.75 m in diameter. To the east and west ofthe mortuary structure, two oval-shaped side chambers weredocumented. The narrow passage has a single segment 1.30 m inlength, with access to the chamber through a perforated sandstoneslab. On the opposite side of the passage, a poorly preservedorthostatic forecourt was also a standing feature. The mound was11 m in diameter with at least 5 concentric stone walls.

The mortuary deposit consisted of a mass of stratified, mixedanthropological remains broadly occupying the central andnorthern part of the main and western chambers. The southern halfof the mortuary area and the corridor were almost completelyempty as a result of looting that has fortunately not affected thewhole tomb. This means that an unknown proportion of theanthropological remains has been lost and consequently the pre-served human bones do not constitute the entire original burialpopulation. This situation is not exceptional and can be found inmost of the graves excavated in this necropolis.

In the undisturbed chamber area, bone remains appear piled ontop of each other, overlapping in many cases. This complex stra-tigraphy was excavated in four layers. Layer I was found 30 cm fromthe surface, Layer II 50 cm, Layer III 70 cm and Layer IV was at thebottom, at a depth of 110 cm and just above the floor of thechamber. This stratigraphy follows a general trend documented inmany other burials in the necropolis (Almagro Gorbea, 1973). It

Fig. 1. Map showing the location of tholoiwith radiocarbon dates in the southeastern Iberia: (1) El Barranquete; (2) Los Millares; (3) La Loma del llano del Jaut�on; (4) La Encantada I.

1 Spanish National Research Council.

G. Aranda Jim�enez, A. Lozano Medina / Journal of Archaeological Science 50 (2014) 369e382370

seems likely that the preserved anthropological remains could berepresentative of the majority of the ritual biography of the tomb. Itis not possible to rule out the possibility that previous intermentshad been cleared out during the Copper Age, although this seemsunlikely, especially if we take into account the fact that no evidenceof this type of ritual behaviour has been reported elsewhere for thesouthern Iberian megalithic phenomenon.

It should be noted that these four layers mainly represent thephase order in which the skeletal remains were excavated. Theregular thickness intervals between the four identified layerswould imply that they were artefacts of excavation and recording.The exception can be found in the differentiation between Layers IIIand IV, in which the associated material culture was the main cri-terion. In fact, Layer III appeared to be related to typical EarlyBronze Age grave goods, in contrast to Layer IV which was linked toChalcolithic pottery vessels (Figs. 4 and 5). The radiocarbon datingfrom 40 years later presented in this paper supports that thedifferent layers identified by the excavators are consistent with achronological succession of depositional events.

The bones of Grave 11 were found in an articulated, semi-articulated and scattered state. These remains could be primarydeposits that were disturbed by later activity in the chambers,mainly subsequent burials. Such activity would have involvedmoving and reorganizing the earlier anthropological remains.Therefore it was assumed that each individual was buried as anarticulated skeleton still with flesh on its bones or at least withbones still connected by ligaments. With the data currentlyavailable it is not possible to rule out the possibility of secondaryburials of bodies that had previously been interred elsewhere andsubsequently placed in Grave 11. Nevertheless, this possibilityappears to be very unlikely if the dating results are taken intoaccount (see below). We also cannot disregard the possibility thatsome bones were removed from the chamber to be depositedelsewhere.

Grave goods are scarce and consist mainly of pottery sherds.Only in one case, that of an adult male found in an anatomicalconnection and flexed position, was it possible to link the remainsto specific grave goods, including a copper dagger, two potteryvessels and a copper bracelet attached to the right radius and ulna.Another copper dagger from Layer I completes the list of metalgrave goods found in this tomb. Of the pottery sherds, differentfragments from the same vessel with Bell-Beaker decoration foundin different areas of the tomb (mound, passage and chamber) canbe highlighted. This is the only vessel with Beaker decoration foundthus far at El Barranquete necropolis. As we will stress below, thisfinding is fully consistent with the radiocarbon measurements.

2. Materials and methods

El Barranquete necropolis was a pioneering site in Iberia interms of radiocarbon dating. At the beginning of the 1970s, foursamples were dated by the CSIC1: two pieces of charcoal from awooden post that supported the roof of the mortuary chamber inGrave 7 and two more of human bones from Grave 11. The samplesfrom Grave 7 produced statistically very consistent radiocarbonmeasurements (CSIC-81 4280 ± 130, 3339e2505 cal BC 2s andCSIC-82 4300 ± 130, 3350e2550 cal BC 2s). Together with the dateof Tomb 19 at the Los Millares necropolis (KN-72, 4380 ± 120,3400e2650 cal BC 2s), they have traditionally been used to date, inbroad terms, the beginning of the tholoi phenomenon in south-eastern Iberia to between the end of the fourth and the begin-ning of the third millennium. Radiocarbon measurements fromGrave 11 revealed a very different picture. The results indicatedvery recent dates (CSIC-201B, 2570 ± 100, 898e412, cal BC 2s andCSIC-201A, 840 ± 100, 995e1301 cal AD 2s); this was considered

Fig. 2. Map of the necropolis of El Barranquete showing the location of megalithic graves (After Almagro Gorbea, 1973).

G.A

randaJim

�enez,A.Lozano

Medina

/Journal

ofArchaeological

Science50

(2014)369

e382

371

Fig. 3. Plan of the grave 11 at the necropolis of El Barranquete (after Almagro Gorbea, 1973).

2 The MNI was calculated for the whole assemblage taking into account the evennumber of bones of the same side of the body. Then sex and age were considered toestablish the MNI of males, females and subadults. A complete study of theanthropological material from Burials 9, 10 and 11 not included in the reportpublished in 1973 is now in progress, and full details will be provided elsewhere.

3 If we take into account the date CSIC-201B, the MNI could reach at least 13individuals.

G. Aranda Jim�enez, A. Lozano Medina / Journal of Archaeological Science 50 (2014) 369e382372

completely inappropriate for the cultural context analysed and itwas therefore assumed that the bone samples were contaminatedor intrusive (Almagro Gorbea, 1973). As a result of the Late BronzeAge radiocarbon dates found at other tholoi, the CSIC-201B date iscurrently acknowledged as a part of the ritual re-use practices ofthese megalithic tombs (Castro et al., 1996; Lorrio and Montero,2004; García Sanju�an et al., 2011; Aranda, 2013).

In 2011, we decided to undertake a radiocarbon dating pro-gramme of different megalithic burials and natural caves in south-eastern Iberia to explore the continuity and reuse of these ritualspaces during the Bronze Age. Our main aim was to assess thescale and involvement attained by this phenomenon in a culturalcontext in which a new funerary ritual consisting of individualinhumations within settlement areas became widespread (Aranda,2013, 2014). As a part of this programme, Grave 11 at El Barran-quete necropolis was selected for dating, as there was materialproof of its reuse during the Early Bronze Age. In fact, of the fourburial layers identified during the excavation process, the firstthree fell within this period (Almagro Gorbea, 1973). Three bonesamples from different individuals were dated; one from Layer IIand two from Layer III. All the radiocarbon measurements fellwithin the Early Bronze Age (Beta-301933, 2190e1940 cal BC 2s,Beta-301934, 1940e1760 cal BC 2s, and Beta-301932,

1890e1690 cal BC 2s), indicating that the permanence and con-tinuity of collective funerary rituals was an outstanding feature(Aranda, 2013, 2014).

Nevertheless, it soon became clear that the number of existingmeasurements was inadequate for providing a detailed under-standing of the chronology of El Barranquete, especially if weintended to use the Bayesian approach to the interpretation ofchronological data. As a result, in 2013 an additional suite of sam-ples was submitted for dating. In this case we planned to date allthe minimum numbers of individuals (MNI) identified in Zita Laf-franchi and Juan Sebasti�an Martín-Fl�orez's anthropological study.Information about sex, age and sidedness was used to avoid repli-cate individuals.2 As a result, the Minimum Number of Individualsidentified was twelve.3 It should be noted that this figure estimatesthe minimum possible number of inhumations, but not the most

Fig. 4. Plan of the grave 11 including the anthropological remains of Layers IeIII (after Almagro Gorbea, 1973).

4 Unfortunately, no bone material belonging to Layer I has been identified amongthe anthropological collection currently deposited in the Museum of Almería. It islikely that there have been some losses in the 40 years since it was excavated.

G. Aranda Jim�enez, A. Lozano Medina / Journal of Archaeological Science 50 (2014) 369e382 373

likely. It is reasonable to expect a higher MNI if the incompletepreservation of the anthropological assemblage is considered. Asimilar situation can be found in the rest of the excavated burials. Infact, the anthropological study of the first eight tombs, which arealso characterized by their incomplete preservation, shows that theMNI ranges from 6 to 23 individuals. Furthermore, Burials 6 and 8offered a similar figure of 12 minimum number of individuals(Botella L�opez, 1973).

The samples from Grave 11 were carefully selected to ensurethat no individuals were dated twice. This pre-condition is veryimportant as the algorithm used in the Bayesian analysis assumesthat every date is statistically independent of the others (BronkRamsey, 2001). Following these criteria, 12 individuals weredated, including samples from adults, infants, males and females(see Table 1 for all the anthropological details). According to thestratigraphic sequence established by the excavators, eight

samples came from Layer IV, two from Layer III and two fromLayer II.4

One relevant concern, especially in cases like El Barranquetenecropolis that are near the Mediterranean Sea, is determining thediet type of the people sampled for dating. If marine resources werea relevant weight in their diet, the radiocarbon measurementscould be strongly influenced by the reservoir effect. The availabledata to assess this issue came from the nearby megalithic necrop-olis of Los Millares, also located near the sea. Stable isotope mea-surements (d13C and d15N) of different individuals and gravesindicate a diet based on plants and terrestrial animals, with very

Fig. 5. Plan of the grave 11 including the anthropological remains of Layer IV (after Almagro Gorbea, 1973).

G. Aranda Jim�enez, A. Lozano Medina / Journal of Archaeological Science 50 (2014) 369e382374

little seafood input.5 Although more analyses are needed, we canassume that the Chalcolithic populations buried in tholoi necrop-olises such as Los Millares or El Barranquete have not been signif-icantly influenced by the reservoir effect, and therefore radiocarbondates from human bone samples can be considered as accurateestimates.

With the exception of Sample CSIC-201B, all recent dates weremeasured using Accelerator Mass Spectrometry and dated by BetaAnalytic Radiocarbon Dating6 and the CNA (National AcceleratorCentre) (Santos Ar�evalo et al., 2009) labs. All results were calibrated

5 Waterman, A.J. and Robert H. T (in preparation) “Evidence of dietary variabilityand breadth at Los Millares (Almeria, Spain) (c. 2500e1800 BC) based on stableisotope analysis of human remains”.

6 Procedure details are described in http://www.radiocarbon.com/.

using the Oxcal (v.4.2) program (Bronk Ramsey, 1995, 2001, 2009),and the IntCal13 calibration curve (Reimer et al. 2013). The datingresults provide a coherent sequence of dates for the megalithicburial (Table 1). The Bayesian models were built using the OxCalprogram, which has been available since 1994 and has subse-quently been developed (Bronk Ramsey, 2013).

3. Objectives

The methodological advances in radiocarbon measurementsand their interpretation in the last few decades (Buck et al., 1991,1999; Bronk Ramsey, 1994, 2013; Taylor, 1997; Bayliss, 2009;Bayliss and Whittle, 2007; Whittle et al., 2011) make it possibleto achieve a much more precise dating of the megalithic phe-nomenon. These developments provide a great opportunity to

Table 1Radiocarbon dates from grave 11 at the necropolis of El Barranquete.

Laboratory code Type of material andanthropological details

Layer Radiocarbonage (BP)

d13C(‰)

Calibrateddate range(95%confidence)Cal BC

First model Second model

Posteriordensityestimate(68%) Cal BC

Posteriordensityestimate(95%) Cal BC

Posteriordensityestimate(68%) Cal BC

Posteriordensityestimate(95%) Cal BC

El Barranquete Tholos 11 dates included in modelsCNA-2405 Human right humerus from an

adult female (21e40 years old)IV 3925 ± 35 �20.19 2558e2295 2383e2292 2469e2233 2384e2292 2468e2236

CNA-2404 Human right radius froman adult female (21e40 years old)

IV 3865 ± 35 �20.10 2465e2209 2367e2231 2431e2207 2368e2232 2432e2208

CNA-2407 Human left radius froman adult male (21e40 years old)

IV 3860 ± 35 �19.72 2463e2208 2355e2215 2432e2206 2357e2230 2433e2207

CNA-2408 Human skull fragment from aninfantile (7e12 years old)

IV 3860 ± 35 �20.22 2463e2208 2354e2215 2431e2206 2357e2230 2433e2207

CNA-2411 Human skull fragment froman adult male (þ30 years old)

IV 3830 ± 35 �19.02 2457e2150 2332e2214 2429e2155 2234e2227 2429e2197

CNA-2410 Human tibia from an adult male(21e40 years old)

IV 3785 ± 35 �25.30 2339e2051 2291e2199 2336e2143 2292e2200 2336e2146

CNA-2406 Human left humerus from anadult male (21e40 years old)

IV 3730 ± 35 �18.82 2276e2028 2286e2164 2297e2115 2286e2169 2295e2196

CNA-2409 Human skull fragment froman adult male (30e35 years old)

III 3625 ± 40 �23.11 2132e1889 2015e1931 2050e1887 2028e1936 2118e1886

Beta-301933 Human skull fragment from anunsexed adult (41e60 years old)

III 3670 ± 40 �18.7 2195e1939 2039e1944 2111e1911 2057e1951 2131e1926

Beta-301934 Human right femur from a adultmale (21e30 years old)

III 3530 ± 30 �18.3 1943e1763 1947e1879 2016e1820 1917e1777 1943e1764

Beta-301932 Human femur from an adultmale (41e60 years old)

II 3470 ± 40 �18.4 1894e1687 1786e1689 1863e1660 1878e1703 1895e1687

CNA-2412 Human right ulna from anadult male (21e40 years)

II 3120 ± 35 �17.89 1492e1285 1490e1324 1498e1303 1490e1326 1498e1303

El Barranquete Tholos 11 date not included in modelsCSIC-201B Human bone e 2570 ± 100 e 898e412 Unmodelled

G. Aranda Jim�enez, A. Lozano Medina / Journal of Archaeological Science 50 (2014) 369e382 375

create a refined chronological framework. Based on the newradiometric data and the stratigraphic interpretation provided byAlmagro Gorbea (1973), alternative models for the chronology ofGrave 11 can now be discussed. The new dating series and theBayesian models were designed to address the followingobjectives:

To determine the chronology of the construction of this sepul-chre and its abandonment as a burial site;

To estimate the absolute date and duration of the burial activityas a whole and the different inhumation layers identified;

To determine if there were gaps between the different inhu-mation levels and, if so, to estimate the length of these hiatus;

To investigate the relevance of the reuse burial practices duringthe Bronze Age and later periods;

To evaluate the Bayesian chronological models of Grave 11 in thecontext of the megalithic phenomenon in southern Iberia;

To place El Barranquete necropolis within its wider regionalsequence.

7 Radiocarbon measurements frequently provide ranges spanning 250 years ormore.

8 Informative prior beliefs in the Bayesian terminology.9 Posterior density estimate in the Bayesian chronology.

4. Results and discussion

Thirteen radiocarbon measurements are now available fromGrave 11, a situation that has never before been achieved for Iberianmegalithic phenomenon. Among the few graves with radiocarbondates, the number of measurements fluctuates between one date,the most common situation, and four, which is only reached at afew sites (Boaventura, 2009; García Sanju�an, 2011; Balsera et al.,forthcoming). Most of these radiocarbon measurements onlyconfirm the Neolithic/Chalcolithic chronology of this cultural phe-nomenon. Therefore, megalithic chronology in Iberia is based ontoo few dates from too few burials.

This current chronometric scenario of the megalithic phenom-enon in southern Iberia imposes major limitations on chronological

studies. Visual inspection of the probability distribution of series ofcalibrated radiocarbon dates is the normal method used to buildchronological models. The most advanced approaches attempt totake into account the statistical scatter on the radiocarbon mea-surements, reducing the probabilistic distribution to a periodwheremost of the dates are concentrated. Nevertheless, this type ofapproach produces imprecise chronologies and broad discussionsthat divide every millennium into no less than three or four in-tervals7 (Bayliss et al., 2007; Whittle and Bayliss, 2007). Bayesiananalysis has emerged as an alternative to build more accuratechronological estimates. This methodology combines evidence ofabsolute chronology, for instance radiocarbon dating, with otherforms of chronological information, such as the stratigraphic re-lationships between the archaeological contexts that contain thedated samples.8 In accordance with these prior beliefs, this tech-nique calculates a shorter probability distribution for every radio-carbonmeasurement and estimates the dates for the beginning andend of the phases or events in which the dates are clustered.9

Furthermore, by comparing these probability distributions, it isalso possible to measure in numbers of years the length of eachphase or the hiatus between periods of social activity (BronkRamsey, 1995; Bayliss et al., 2007).

The Bayesian approach is causing a profound change in ourperception of temporality in prehistoric societies. Where thenumber and diversity of samples that can potentially be dated issufficient and the stratigraphic or sequential information is alsoaccurate, the chronological models have demonstrated a

Fig. 6. Probability distributions of dates from grave 11. Each date shows two distributions: one in light grey represents the radiocarbon calibration and another in dark grey indicatesthe result of the bayesian model (posterior density estimates). Distributions other than those relating to particular dates correspond to aspects of the model. The square bracketsdown the left-hand side and the OxCal keywords define the overall model exactly.

G. Aranda Jim�enez, A. Lozano Medina / Journal of Archaeological Science 50 (2014) 369e382376

misleading impression of longevity in some cultural phenomenaand the possibility of reducing the broad spans of prehistoricchronology to the scale of an individual lifetime (Whittle andBayliss, 2007; Whittle et al., 2008, 2011; Scarre, 2010). As has

Fig. 7. Probability distributions of the number of years for every ph

recently been stated, “It is now possible to think in terms of life-times, lifespan and generations” (Whittle and Bayliss, 2007: 23).

In the case study analysed in this paper, the conditions are farfrom ideal, with several existing limitations that affect the degree of

ase of burial activity, derived from the model shown in Fig. 5.

G. Aranda Jim�enez, A. Lozano Medina / Journal of Archaeological Science 50 (2014) 369e382 377

resolution that can be achieved for the model. The only materialavailable for dating is the anthropological remains; these have alow Minimum Number of Individuals and an uneven distributionbetween the different phases identified through the inhumationsequence of the burial. Furthermore, the archaeological informa-tion does not allow the stratifying of mortuary depositions withineach phase. The limited number of suitable samples and theparticular stratigraphic relationships reduce the options forproviding robust answers to some of the objectives of this study.Despite these limitations, Bayesian analysis certainly remains thebest option for building a detailed and more secure chronology andfor improving the chronological assessments that do not take intoaccount the statistical dispersion of the radiocarbon dating prob-ability intervals.

Therefore, the main relative chronological information used toconstruct the Bayesian model is the four stratified layers identifiedin the burial sequence. This arrangement of the bone remains hasnot taken into account the complex taphonomy that normallyoperates in such contexts. Horizontal and vertical displacements,both of individual and grouped bones, are common as a result ofdifferent factors such as gravity, successive interments, voidscreated by the decomposition of soft tissue, or the collapse of the

Fig. 8. Probability distributions of dates from burial 11, incorporating the inte

burial chamber vaults. The Bayesian modelling incorporates a sta-tistic known as the index of agreement, which calculates the reli-ability of the model and provides useful information to identifysamples whose archaeological taphonomy has not been properlycharacterized. This index estimates a figure of how well any pos-terior probability distribution agrees with the relative sequenceinformation (Bronk Ramsey, 1995: 427e8). If the index of agree-ment falls below 60%, the radiocarbon measurement should beconsidered somewhat problematic.

Out of the twelve radiocarbon dates, only one (CNA-2409) pre-sents a poor agreement index A ¼ 47.1% (�Ac ¼ 60%). This sampleappears to be affected by taphonomic processes that would imply avertical movement. Although according to the excavators thesample belongs to the fourth layer, we have assumed it is part of theinterments above. Furthermore, the posterior density estimates arenot substantially modified if the CNA-2409 date is included as partof the fourth or the third level of inhumations.

Taking into account the available archaeological informationand its restrictions, we built two chronological models that incor-porate two assumptions: a) that within each phase, samplescorrespond to a continuous activity in which the internal order isunknown and b) that all the individuals died after the tholos had

rpretation of the second model. The format is identical to that in Fig. 5.

Fig. 9. Probability distributions of the number of years for every phase of burial activity, derived from the model shown in Fig. 7.

G. Aranda Jim�enez, A. Lozano Medina / Journal of Archaeological Science 50 (2014) 369e382378

been built (see below for further discussion). According to thestratigraphic and sequential information, the first model includesseven dates in Layer IV, three in Layer III and just two in Layer II(Fig. 6 and Table 1). This model displays good overall agreement

Table 2Radiocarbon dates from the tholoi-type graves of southern Iberia.

Site Province/Region Laboratorycode

Type of material Contex

Cabeço da Arruda 2 Lisboa (Portugal) Sac-1613 Human left femur e

Huerta Montero Badajoz (Spain) GrN-17628 Human bone First deat the c

Castro Marim Algarve (Portugal) OxA-5441 Human tibia e

Cabeço da Arruda 2 Lisboa (Portugal) Sac-1784 Human left femur e

Praia das Maç~as Lisboa (Portugal) OxA-5509 Bone Occiden

Praia das Maç~as Lisboa (Portugal) OxA-5510 Bone Occiden

Los Millares Almería (Spain) KN-72 Wood charcoal ofmixed composition

Grave X

El Barranquete Almería (Spain) CSIC-82 Charcoal Grave 7

Olival da Pega 2b Alentejo (Portugal) ICEN-955 Human bone Phase 1Deposi

El Barranquete Almería (Spain) CSIC-81 Charcoal Grave 7

Praia das Maç~as Lisboa (Portugal) H-2049/1467 Charcoals Occiden

Paimogo 1 Lisboa (Portugal) Sac-1556 Human left femur e

Cabeço da Arruda 2 Lisboa (Portugal) UBAR-538 Human left femur e

Huerta Montero Badajoz (Spain) GrN-16955 Human bone Secondor reusat the c

Olival da Pega 2b Alentejo (Portugal) ICEN-956 Burnt bonesand charcoals

Phase 1Deposi

La Pijotilla Badajoz (Spain) CNA-034 Charcoal Grave 3(lower

La Loma del llanodel Jaut�on

Almería (Spain) ¿ Human bone Chamb

Paimogo 1 Lisboa (Portugal) UBAR-539 Human left femur e

Olival da Pega 2b Alentejo (Portugal) ICEN-957 Burnt bonesand charcoals

Phase 1Deposi

La Pijotilla Badajoz (Spain) Beta-121143 Charcoal Grave 3(interm

Agualva Lisboa (Portugal) Beta-239754 Human left femur e

Paimogo 1 Lisboa (Portugal) Sac-1782 Human left femur e

Tintuaria Lisboa (Portugal) OxA-5446 Human bones (?) e

Samarra Lisboa (Portugal) Sac-1827 Human bone e

Huerta Montero Badajoz (Spain) GrN-16954 Human bone Secondreuse a

Praia das Maç~as Lisboa (Portugal) H-2048/1458 Charcoals Chamb

La Encantada I Almería (Spain) CSIC-249 Human bone Chamb

Tholos de Palacio III Sevilla (Spain) Beta-165552 Charcoal Structu

(Aoverall ¼ 74.2%), which indicates that the radiocarbon datesconform to the archaeological information incorporated in theBayesian analysis. If we accept that the earlier interments weremade just after the construction of the tholoi had been completed, it

t Radiocarbonage (BP)

Calibrateddate range(68%confidence)Cal BC

Calibrated daterange (95%confidence)Cal BC

Bibliography

4700 ± 80 3629e3373 3652e3196 Silva, 2012posithamber

4650 ± 250 3656e3026 3981e2697 Blasco and Ortiz,1991

4525 ± 60 3356e3106 3491e3024 Gomes et al.,1994

4460 ± 45 3328e3028 3347e2942 Boaventura, 2009tal chamber 4410 ± 75 3312e2917 3339e2903 Cardoso and

Soares, 1995tal chamber 4395 ± 60 3097e2916 3331e2899 Cardoso and

Soares, 1995IX 4380 ± 120 3325e2894 3485e2678 Schwabedissen

and Freundlich,1966

. Wooden post 4300 ± 130 3263e2677 3341e2581 Almagro Gorbea,1973

t level 5/74290 ± 100 3090e2699 3330e2586 Soares, 1999;

Gonçalves, 2006. Wooden post 4280 ± 130 3092e2670 3339e2505 Almagro Gorbea,

1973tal chamber 4260 ± 60 3000e2701 3077e2638 Soares and

Cabral, 19844250 ± 90 3009e2671 3261e2574 Boaventura, 20094230 ± 100 2923e2632 3096e2495 Silva, 2012

depositehamber

4220 ± 100 2916e2632 3090e2495 Blasco and Ortiz,1991

t level 7/74180 ± 80 2887e2666 2920e2497 Soares, 1999;

Gonçalves, 2006. UE 18level)

4168 ± 55 2877e2677 2891e2583 Odriozola Lloretet al., 2008

er 4160 ± 60 2876e2668 2889e2581 Maicas Ramos,2007

4130 ± 90 2871e2584 2896e2486 Silva, 2012

t level 5/74130 ± 60 2866e2620 2884e2501 Soares, 1999;

Gonçalves, 2006. UE 15ediate level)

4130 ± 40 2862e2625 2872e2581 Hurtado P�erezet al., 2002

4110 ± 40 2855e2582 2871e2505 Boaventura, 20094100 ± 60 2859e2574 2876e2491 Silva, 20123995 ± 65 2624e2369 2851e2297 Cardoso et al.,

19963820 ± 60 2401e2146 2466e2060 Silva et al., 2006.

deposit ort the chamber

3720 ± 100 2284e1974 2458e1889 Blasco and Ortiz,1991

er 3650 ± 60 2133e1942 2201e1884 Soares andCabral, 1984

er 2830 ± 60 1074e906 1192e837 Alonso Mathiaset al., 1978

re 3 (cremation) 2660 ± 90 971e766 1043e540 García Sanju�an,2005a

G. Aranda Jim�enez, A. Lozano Medina / Journal of Archaeological Science 50 (2014) 369e382 379

is possible to conclude that the tomb was finished almost at thesame time as the first bodies were deposited in 2551e2242 cal BC(95% probability; Boundary_Start 1), probably in the second half ofthe twenty-fifth or during the twenty-fourth century(2452e2316 BC: 68% probability; Boundary_Start 1). The end of thisPhase IV, which could be considered the proper Chalcolithic periodof this monument, occurred in 2283e2054 cal BC (95% probability;Boundary_End 1), probably in the second half of the twenty-third orfirst half of the twenty-second century (2256e2128 BC: 68% prob-ability; Boundary_End 1). This probability distribution is fullyconsistent with the date c. 2200 cal BC, which is traditionallyconsidered as a benchmark for the end of the Copper Age and thebeginning of the Early Bronze Age (known as the Argaric Culture insouth-eastern Iberia) (Aranda Jim�enez et al., forthcoming). Thedifferences between these two distributions suggest that themortuary deposition continued for between 81 and 249 years (68%probability; Span_Layer IV) (Fig. 7). The dates included in this phasehave not provided statistically consistent measurements (T0¼ 19.4;T0 (5%) ¼ 12.6) (Ward and Wilson, 1978), which also means thatpeople must have died and been deposited in the burial chamberover a period of time.

The burial activity in the second phase (Layer III) began in2185e1930 cal BC (95% probability; Boundary_Start 2), probably inthe twenty-first or the first half of the twentieth century(2100e1966 cal BC: 68% probability; Boundary_Start 2). It appearsthat there was no hiatus between the two phases, or if there were,it was very short. The radiocarbon measurements for the mostrecent date from Layer IV (CNA-2406) and the earliest date fromLayer III (Beta-301933) are statistically consistent (T0 ¼ 1.3; T0

(5%) ¼ 3.8), suggesting that these people died over a relativelyshort period of time. Although with less intensity, the continuity inthe use of the grave appears to be an outstanding feature. The endof this phase occurred in 1978e1760 cal BC (95% probability;Boundary_End 2), probably between the last decades of the twen-tieth century and the first half of the nineteenth century(1937e1838 cal BC: 68% probability; Boundary_End 2). At the sametime as this phase ended, major changes in rituals were takingplace in the Argaric societies. The funerary rights were extended toother groups of people, especially children, and a new set of gravegoods, including swords, axes and diadems, appeared for the firsttime (Lull et al., 2011). As with Layer IV, in this phase the radio-carbon measurements do not form a statistically consistent group(T0 ¼ 8.8; T0 (5%) ¼ 6.0). Again the funerary deposits were probablymade over a relatively long period lasting approximately 217 years(95% probability, Span_Layer IV).

The limited number of measurements available for the thirdphase (Layer II) has resulted in imprecise probability estimates. Thebeginning of this phase occurred in 1921e1695 cal BC (95% proba-bility; Boundary_Start 3), probably between the nineteenth centuryand the first half of the eighteenth century (1873e1747 cal BC: 95%probability; Boundary_Start 3). The last interments in the burialchamber took place in 1493e951 cal BC (95% probability; Boundar-y_End 3), probably between 1430 and 1230 cal BC (68% probability;Boundary_End 3). It should be noted that the CSIC-201B date has notbeen modelled, as the context and stratigraphic position of thesample are unknown. If the results of this date were taken intoaccount, the period of use would extend into the first millenniumcal BC.

A second model was built combining Layers III and II in a singlephase. According to the excavators, these layers can be clusteredbecause they contain grave goods that place them typologically inthe Early Bronze Age (Fig. 8). The model that only includes twophases (Layer IV and Layers IIIeII) also has a good overall agree-ment (Aoverall¼ 83.9%), indicating that the radiocarbon dates do notdisagree with the prior information. The posterior density

estimates for the dates provided by this second model are verysimilar to those given by the first model (Table 1). The main changecan be found at the beginning of the second phase (Layers IIIeII),which occurred in 2218e1974 cal BC (95% probability; Boundar-y_Start 2), probably between the second half of the twenty-secondand the first half of the twenty-first century (2154e2022 cal BC: 68%probability; Boundary_Start 2). According to this model, the in-terments in this phase would have started a little earlier than in theprevious model. These probability estimates are even closer to thedate c. 2200 cal BC, which is conventionally acknowledged as thebeginning of the Early Bronze Age.

This second phase (Layer IIIeII) spans a period of between 493and 769 years (95% probability, Span_Layer IIIeII) or more likelybetween 552 and 629 years (68% probability, Span_Layer IIIeII)(Fig. 9). These figures indicate that the mortuary chamber was inuse for a very long time, almost the entire Bronze Age. This scenariocontrasts strongly with the shorter period inwhich the Chalcolithicphase of the tomb (Layer IV) was in use: between 17 and 302 years(95% probability, Span_Layer IV) and more likely between 75 and239 years (68% probability, Span_Layer IV). If we assume a figure of25 years for each generation, just a few generations e betweenthree and nine e were buried in this tholos during the Copper Age.Both models display good overall agreement and posterior densityestimates are very similar, with only slight differences. Now wemove on to discuss all these results in a more general culturalcontext.

5. Conclusions

Using the dating results presented above, we can now note anumber of broad suggestions as to howGrave 11 and the necropolisof El Barranquete fit into the local and regional sequence. To date ithas not been possible to determine the timescale and funerary spanof any megalithic monument, due to the lack of complete datingprogrammes for single graves. Despite the sample restrictionspreviously noted, we now have a much more precise chronology,which raises fresh questions and shakes traditional assumptions.

The Bayesian models underline how late the construction anduse of Grave 11 is. If we assume that interments began in the burialchamber just after its construction, this grave would have beenbuilt in the second half of the twenty-fifth or the first half of thetwenty-fourth century cal BC (2452e2316 BC, 68% probability). As afar as we know, it is one of the latest megalithic monuments to beconstructed in southern Iberia. By the time the first bodies weredeposited at this burial site, other tholoi would have already beenolder and theymay even no longer have been used for ritual activity(Table 2).

The beginning of the construction and use of this tholoi-typetomb has been traditionally dated to the last centuries of thefourth millennium, as a part of the second megalithic tradition(García Sanju�an, 2011). The oldest radiocarbon dates currentlyknown in southern Iberia correspond to different Portuguesetholoi, such as Praia Das Maças (OxA-5509, 3312e2917 cal BC2s), Paimogo 1 (Sac-1556, 3261e3098 cal BC 2s), Cabeço daArruda 2 (Sac-1613, 3652e3196 cal BC 2s) and Castro Marim(OxA-5441, 3490e3020 cal BC 2s). From the end of the fourthmillennium and mainly during the first centuries of the thirdmillennium, this new type of megalithic tomb became wide-spread throughout southern Iberia. In this context, the notionthat the tholoi monuments remained in use over a span ofcenturies containing long sequences of mortuary depositionsbecame widely accepted.

Nevertheless, the late construction of Grave 11 challenges theseassumptions. In Fig. 10 we have joined the Bayesian model with thedates of the southern Iberian tholoi graves, which produces a

Fig. 10. Relationship between the Bayesian model of grave 11 shown at the bottom and all the tholoi radiocarbon dates available for the south of Iberia.

10 The large standard deviations of dates from Grave 7 make it impossible toprovide a more accurate probability.

G. Aranda Jim�enez, A. Lozano Medina / Journal of Archaeological Science 50 (2014) 369e382380

temporal gap of no less than five centuries between Grave 11 andmost of the dates. For a more local perspective, the El Barranqueteradiocarbon dates also indicate how the burials of a single ne-cropolis could have been temporarily sequenced. As we havealready noted above, Tomb 7 has provided two dates from thewooden post that supported the roof (CSIC-81 4280 ± 130,3339e2505 cal BC 2s and CSIC-82 4300 ± 130, 3350e2550 cal BC2s). The combination of these dates (R_combine 4290 ± 92,3327e2621 cal BC 2s, 3087e2702 cal BC 1s) confirms that itsconstruction and use probably began at the very end of fourthmillennium or during the first centuries of the third millennium.This indicates that Grave 7 was finished between seven and threecenturies before the first inhumations were deposited in Grave 11

(338e683 year at 68% or 190e905 at 95%, Difference constructionBarranquete 11 and 7).10

The late construction of Grave 11 also challenges the idea of longsequences of inhumations that would roughly occupy the wholeChalcolithic period or at least most of it. During this phase (LayerIV), the ritual use of the funerary chambers was intense but short,between 75 and 239 years (68% probability), which would meanbetween three and nine generations of successive mortuary de-positions. All these temporal differences in the construction and

G. Aranda Jim�enez, A. Lozano Medina / Journal of Archaeological Science 50 (2014) 369e382 381

use of the megalithic monuments indicate the weaknesses of thosesocial interpretations that consider this phenomenon as tempo-rarily unitary (Afonso Marrero et al., 2011). The internal chronologyand especially the diachrony of megalithic necropolises such as ElBarranquete or Los Millares emerge as crucial factor for a betterunderstanding of these social practices.

A final implication of the late construction of Grave 11 is relatedto the possibility that ancestral remains were deposited within theburial chamber. Because the skeletons appear in varying states ofarticulation and completeness, this option remains open. Never-theless, the presence of bones significantly older than those foundin Grave 11 places the construction date of the tholos in the EarlyBronze Age, which, according to our current knowledge, appears tobe quite unlikely. In fact, the short period of time that the firstphase of the mortuary structure was in use does not leave room forassuming that ancestral remains were placed in a tomb built duringthe Chalcolithic period.

One of the most outstanding features of Grave 11 is its “reuse”after the Chalcolithic phase (Layers III and II). According to theBayesian models, the beginning of these ritual practices occurred in2218e1974 cal BC at 95% probability or in 2154e2022 cal BC at 68%probability, which agrees with the beginning of the Early BronzeAge. Statistically, there are no chronological differences with theChalcolithic interments, suggesting that there was continuity in theuse of this ritual space without any major hiatus. This is quite aremarkable phenomenon and the lack of any temporal break hasled us to wonder if the traditional concept of reuse is appropriatefor this type of ritual behaviour. Is it possible that before c.2200 cal BC we are dealing with a “use phase” and after that datewith a “reuse phase”? The radiocarbon dates from Grave 11 chal-lenge this notion, showing continuous mortuary depositions dur-ing the second half of the thirdmillennium and the first centuries ofthe second millennium cal BC. The benchmark of c. 2200 cal BCseems to work for other cultural manifestations, but not for thistype of funerary ritual. The phenomena of continuity and discon-tinuity shows a complex scenario of cultural manifestations and theinadequacy of traditional Copper and Bronze Age divisions.

Although the mortuary depositions related to this phase of usespan a very long period (552e629 years at 68% probability), theirfrequency remains a significant feature. If the Minimum Number ofIndividuals is considered as representative of the temporal distri-bution of inhumations, c. 40% of the individuals buried in Grave 11would correspond to the Bronze Age. This figure is consistent withthe situation in south-eastern Iberia. Among the total of 18 knowntholoi grave dates, seven (c. 38%) belong to more recent culturalperiods than those considered in terms of construction and use.

Because of the low number of dates, these figures could not beassumed to be indicative of the relative weight of the “reuse”practices for a wide regional perspective. Nevertheless, the evi-dence of this phenomenon is growing quickly for the study area(Lorrio and Montero, 2004; García Sanju�an, 2005b, 2011; Mataloto,2007; Aranda et al. 2013). In this respect, it is especially noticeablethat the frequent occurrence of radiocarbon dates for human bonesthat fall into cultural periods would be unthinkable if only the ty-pology of their grave goods were analysed (Aranda, 2013, 2014).This is also the case of Grave 11, where there is no archaeologicalevidence that would lead one to suspect the presence of intermentsfrom the times shown by the dates CNA-2412 (3120 ± 35,1492e1285 cal BC 2s) and CSIC-201B (2570 ± 100, 898e412 cal BC2s). It is true that there are not many dates, although thosewe haveallow us to glimpse a panorama involving the continuity of certainritual spaces on a scale and intensity that, according to our currentstate of knowledge, is hard to imagine.

The dating project presented in this paper, which is based on asingle grave instead of several dates from different burials, emerges

as a powerful tool that contributes to changing many of our currentapproaches to the megalithic phenomenon. Grave 11 displays ashort but intense ritual use during the Copper Age, followed by amuch longer period with the interments taking place over manygenerations. Despite the previously noted limitations of theradiocarbon dating, the results of the Bayesian models allow theconstruction of much more precise chronologies that challengedifferent traditional assumptions and open the way for unexpectedresearch inquiries. New dating programmes are a critical necessityif we really want to base our assessments on robust chronologicalfoundations and if our goal is to discuss the agency of particularpeople within a human lifespan.

Acknowledgements

The authors would like to thank Leonardo García Sanju�an andMargarita S�anchez Romero for their comments on early drafts ofthis paper and to the reviewers for their helpful comments. Wewould also like to thank the Almería Museum staff for their sup-port. This paper has been written within the research group “GEA.Material Culture and Social Identity in the Late Prehistory ofSouthern Iberia (HUM-065)”.

References

Afonso Marrero, J.A., C�amara Serrano, J.A., Martínez Fern�andez, G., MolinaGonz�alez, F., 2011. Objetos en materias primas ex�oticas y estructura jer�arquicade las tumbas de la necr�opolis de Los Millares (Santa Fe de Mondújar, Almería,Espa~na). In: García Sanju�an, L., Scarre, C., Wheatley, D.W. (Eds.), Exploring Timeand Matter in Prehistoric Monuments: Absolute Chronology and Rare Rocks inEuropean Megaliths, Proceedings of the 2nd European Megalithic StudiesGroup Meeting (Seville, Spain, November 2008). Menga: Rev. Prehist. Andal.Monogr. 1. Consejería de Cultura de la Junta de Andalucía, Sevilla, pp. 295e317.

Almagro Gorbea, M.J., 1973. El poblado y la necr�opolis de El Barranquete (Almería).In: Acta Arqueol�ogica Hisp�anica VI. Ministerio de Educaci�on y Cienca, Madrid.

Alonso Mathias, J., Cabrera Vald�es, V., Chapa Brunet, T., Fern�andez-Miranda, M.,1978. �Indice de fechas arqueol�ogicas de C-14 para Espa~na y Portugal. In:Almagro Gorbea, M., Fern�andez-Miranda, M. (Eds.), C14 y Prehistoria de laPenínsula Ib�erica. Fundaci�on Juan March, Madrid, pp. 155e183.

Aranda Jim�enez, G., 2013. Against Uniformity cultural diversity: the “Others” inargaric Societies. In: Cruz Berrocal, M., García Sanju�an, L., Gilman, A. (Eds.), ThePrehistory of Iberia. Debating Early Social Stratification and the State. Rout-ledge, New York, pp. 99e118.

Aranda Jim�enez, G., 2014. La memoria como forma de resistencia cultural. Con-tinuidad y reutilizaci�on de espacios funerarios colectivos en �epoca arg�arica.Movilidad, Contacto y Cambio. In: II Congreso de Prehistoria de Andalucía(Antequera, Espa~na 2012). Consejería de Educaci�on, Cultura y Deporte de laJunta de Andalucía, Sevilla.

Aranda Jim�enez, G., García Sanju�an, L., Lozano Medina, A., Costa Caram�e, M.E., 2013.Nuevas dataciones radiom�etricas del dolmen de Viera (Antequera, M�alaga). LaColecci�on G�omez-Moreno. Menga. Rev. Prehist. Andal. 4, 235e249.

Aranda Jim�enez, G., Mont�on Subías, S., S�anchez Romero, M., 2014. The Archaeologyof Bronze Age Iberia: Argaric Societies. Routledge, New York (forthcoming).

Balsera Nieto, V., Bernabeu Aub�an, J., Costa Caram�e, M., Díaz-del-Río, P., GarcíaSanju�an, L., Pardo Gord�o, S., 2014. The radiocarbon chronology of southernSpain's Late Prehistory (5600e1000 cal BC): a comparative review. Oxf. J.Archaeol. 34 (2) (forthcoming).

Bayliss, A., 2009. Rolling out revolution: using radiocarbon dating in archaeology.Radiocarbon 51 (1), 123e177.

Bayliss, A., Whittle, A. (Eds.), 2007. Histories of the Dead: Building Chronologies forFive Southern British Long Barrows. Camb. Archaeol. J. 17 (Special Issue).

Bayliss, A., Bronk Ramsey, C., van der Plicht, J., Whittle, A., 2007. Bradshaw andBayes: towards a Timetable for the Neolithic. Camb. Archaeol. J. 17 (1), 1e28.http://dx.doi.org/10.1017/S0959774307000145.

Blasco, F., Ortiz, M., 1991. Trabajos arqueol�ogicos en Huerta Montero (Almendralejo,Badajoz). In: Actas de las I Jornadas de Prehistoria y Arqueología en Extrem-adura (1986e1990) (M�erida-C�aceres, Espa~na). Extremadura Arqueol�ogica II,pp. 129e138.

Boaventura, R., 2009. As antas e o Megalitismo da regi~ao de Lisboa. Doctoral Thesis.Universidad de Lisboa.

Botella L�opez, M.C., 1973. Estudio de los huesos humanos. El poblado y la necr�opolisde El Barranquete (Almería). In: Acta Arqueol�ogica Hisp�anica VI. Ministerio deEducaci�on y Cienca, Madrid, pp. 229e241.

Bronk Ramsey, C., 1994. Analysis of chronological information and radiocarboncalibration: the program OxCal. Archaeol. Comput. Newsl. 41, 11e16.

Bronk Ramsey, C., 1995. Radiocarbon calibration and analysis of stratigraphy: theOxCal program. Radiocarbon 37 (1), 425e430.

G. Aranda Jim�enez, A. Lozano Medina / Journal of Archaeological Science 50 (2014) 369e382382

Bronk Ramsey, C., 2001. Development of the radiocarbon calibration program.Radicarbon 43 (2), 355e363.

Bronk Ramsey, C., 2009. Bayesian analysis of radiocarbon dates. Radiocarbon 51 (1),337e360.

Bronk Ramsey, C., 2013. Recent and planned developments of the program OxCal.In: Proceedings of the 21st International Radiocarbon Conference. Radiocarbon55 (2e3), pp. 720e730. http://dx.doi.org/10.2458/azu_js_rc.55.16215.

Buck, C.E., Kenworthy, J., Litton, C.D., Smith, A.F.M., 1991. Combining archaeologicaland radiocarbon information: a Bayesian approach to calibration. Antiquity 65,808e821.

Buck, C.E., Christen, J.A., James, G.N., 1999. BCal: an on-line Bayesian radiocarboncalibration tool. Internet Archaeol. 7, 1e6.

Cardoso, J.L., Soares, A.M., 1995. Sobre a cronologia absoluta das grutas artificiais daEstremadura portuguesa. Al-madan 2 (4), 10e13.

Cardoso, J.L., Leit~ao, M., Ferreira, O., North, C., Norton, J., Medeiros, J., Sousa, P., 1996.O monumento pr�e-hist�orico de Tituaria, Moinhos da Casela (Mafra). Estud.Arqueol. Oeiras 6, 135e193.

Castro, P., Lull, V., Mic�o, R., 1996. Cronología de la Prehistoria Reciente de laPenínsula Ib�erica y Baleares (c. 2800e900 cal ANE). In: BAR International Series,vol. 652. Oxford.

García Sanju�an, L., 2005a. Grandes Piedras Viejas, Memoria y Pasado. Reutiliza-ciones del Dolmen de Palacio III (Almad�en de la Plata, Sevilla) durante la Edaddel Hierro. El Periodo Orientalizante. In: Celestino P�erez, S., Jim�enez �Avila, J.(Eds.), El Periodo Orientalizante. Actas del III Simposio Internacional deArqueología de M�erida. Protohistoria del Mediterr�aneo Occidental (M�erida,Espa~na, mayo 2003), Anejos del Archivo Espa~nol de Arqueología 35. CSIC,Madrid, pp. 595e604.

García Sanju�an, L., 2005b. Las piedras de la memoria. La permanencia del mega-litismo en el suroeste de la Península Ib�erica durante el II y I milenios ANE. Trab.Prehist. 62 (1), 85e109.

García Sanju�an, L., 2011. Transformations, invocations, echoes, resistance: theassimilation of the past in Southern Iberian Monumental Landscapes (V to Imillennia BC). In: Lillios, K. (Ed.), Comparative Archaeologies. The AmericanSouthwest (AD 900e1600) and the Iberian Peninsula (3000e1500 BC). OxbowBooks, Oxford and Oakville, pp. 81e102.

García Sanju�an, L., Wheatley, D.W., Costa Caram�e, M.E., 2011. La cronologíanum�erica del fen�omeno megalítico en el Sur de Espa~na: avances y problemas.In: García Sanju�an, L., Scarre, C., Wheatley, D.W. (Eds.), Exploring Time andMatter in Prehistoric Monuments: Absolute Chronology and Rare Rocks inEuropean Megaliths, Proceedings of the 2nd European Megalithic StudiesGroup Meeting (Seville, Spain, November 2008). Menga: Rev. Prehist. Andal.Monogr. 1. Consejería de Cultura de la Junta de Andalucía, Sevilla,pp. 121e142.

Gomes, M.V., Cardoso, J.L., Cunha, A.S., 1994. A sepultura de Castro Marim. Comun.Inst. Geol. Min. 80, 99e105.

Gonçalves, V.S., 2006. Quelques questions autour du temps, de l'espace et dessymboles m�egalithiques du centre et sud du Portugal. In: Joussaume, R.,Laporte, R., Sacarre, C. (Eds.), Origin and Development of the Megalithic Phe-nomenon of Western Europe, Proceedings of the International Symposium(Bougon, France, October 2002). Conseil G�en�eral de Deux S�evres, Niort,pp. 485e510.

Hurtado P�erez, V., Mondejar Fern�andez de Quincoces, P., Pecero Espín, J.C., 2002.Excavaciones en la Tumba 3 de La Pijotilla. In: Jim�enez �Avila, J.J., EnríquezNavascu�es, J.,J. (Eds.), El Megalitismo en Extremadura. Homenaje a Elías Di�eguezLuengo, Extremadura Arqueol�ogica VIII, pp. 249e266.

Lorrio Alvarado, A.J., Montero Ruiz, I., 2004. Reutilizaci�on de sepulcros colectivos enel sureste de la Península Ib�erica: la colecci�on Siret. Trab. Prehist. 61 (1),99e116.

Lull, V., Mic�o, R., Rihuete Herrada, C., Risch, R., 2011. The Argar and the beginning ofClass Society in the Wetern Mediterranean. In: Hansen, S., Müller, J. (Eds.),Sozialarch€aologische Perspektiven: Gesellschaftlicher Wandel 5000e1500 v.Chr. Zwischen Atlantik und Kaukasus. Deutsche Arch€aeologisches Institut-VonZabern, Berlin, pp. 381e414.

Maicas Ramos, R., 2007. Industria �osea y funcionalidad: Neolítico y Calcolítico en laCuenca de Vera (Almería). Consejo Superior de Investigaciones Científicas,Madrid.

Mataloto, R., 2007. Paisagem, memoria e identidade: tumulaçoes megalíticas nop�os-megalitismo alto-alentejano. Rev. Port. Arqueol. 10 (1), 123e140.

Odriozola Lloret, C., Hurtado P�erez, V., Dias, M.I., Prudencio, M.I., 2008. Dataci�on port�ecnicas luminiscentes de la tumba 3 y el conjunto campaniforme de la Pijotilla(Badajoz, Espa~na). In: VII Congreso Ib�erico de Arqueometría. Instituto de His-toria (CSIC), Museo Arqueol�ogico Nacional y SAPaC, Madrid, pp. 211e225.

Reimer, P.J., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Bronk Ramsey, C.,Grootes, P.M., Guilderson, T.P., Haflidason, H., Hajdas, I., Hatt�Z, C., Heaton, T.J.,Hoffmann, D.L., Hogg, A.G., Hughen, K.A., Kaiser, K.F., Kromer, B., Manning, S.W.,Niu, M., Reimer, R.W., Richards, D.A., Scott, E.M., Southon, J.R., Staff, R.A.,Turney, C.S.M., van der Plicht, J., 2013. IntCal13 and Marine 13 radiocarbon agecalibration curves 0e50,000 years cal BP. Radiocarbon 55 (4), 1869e1887.http://dx.doi.org/10.2458/azu_js_rc.55.16947.

Santos Ar�evalo, F.J., G�omez Martínez, I., García Le�on, M., 2009. Radiocarbon mea-surement program at the Centro Nacional de Aceleradores (CNA), Spain.Radiocarbon 51 (2), 883e889.

Scarre, C., 2010. Rocks of ages: tempo and time in megalithic monuments. Eur. J.Archaeol. 13 (2), 175e193. http://dx.doi.org/10.1177/1461957110370731.

Schwabedissen, H., Freundlich, J., 1966. K€oln radiocarbon measurements I. Radio-carbon 8, 239e247.

Silva, A.M., 2012. Antropologia Funer�aria e Paleobiologia das Populaç~oes Portuguesas(Litorais) do Neolítico Final/Calcolítico. Fundaç~ao Calouste Gulbenkian, Lisboa.

Silva, A.M., Ferreira, M.T., Codinha, S., 2006. Praia da Samarra: an�alise antropol�ogicados restos �osseos humanos depositados no Museu Arqueol�ogico de S~ao Miguelde Odrinhas. Rev. Port. Arqueol. 9 (2), 157e169.

Soares, A.M., 1999. Megalitismo e Cronologia Absoluta. In: Balbín Belrmann, R. de,Bueno Ramírez, P. (Eds.), II Congreso de Arqueología Peninsular 3, PrimerMilenio y Metodología. Fundaci�on Rei Afonso Henriques, Zamora, pp. 689e706.

Soares, A.M., Cabral, J.M.P., 1984. Datas convencionais de radiocarbono para estaç~oesarqueol�ogicas portuguesas e a sua calibraç~ao: revis~ao crítica. O Arque�ol. Port. 4(2), 167e214.

Taylor, R.E., 1997. Radiocarbon dating. In: Taylor, R.E., Aitken, M.J. (Eds.), Chrono-metric Dating in Archaeology. Springer Science, Business Media, LLC, New York,pp. 65e96.

Ward, G.K., Wilson, S.R., 1978. Procedures for comparing and combining radio-carbon age determinations: a critique. Archaeometry 20, 19e31.

Whittle, A., Bayliss, A., 2007. The times of their lives: from chronological precisionto kinds of history and change. Camb. Archaeol. J. 17 (1), 21e28. http://dx.doi.org/10.1017/S0959774307000030.

Whittle, A., Bayliss, A., Healy, F., 2008. The timing and tempo of change: examplesfrom the fourth millennium cal. BC in Southern England. Camb. Archaeol. J. 18(1), 65e70. http://dx.doi.org/10.1017/S0959774308000061.

Whittle, A., Healy, F., Bayliss, A., 2011. Gathering Time: Dating the Early NeolithicEnclosures of Southern Britain and Ireland. Oxbow, Oxford.